Walking ability determination device, walking ability determination method, walking ability determination program, model generation device, model generation method, and model generation program

Abstract

To appropriately evaluate walking ability of an object person by a mechanical method.SOLUTION: A walking ability determination device acquires object data generated by observing walking of an object person, and determines walking ability of the object person on the basis of the acquired object data using a trained determination model generated by machine learning. The object data is configured so as to include at least one of first information indicating an area on the coronal plane of a trajectory of a human body's center of gravity at the time of walking of the object person, second information indicating an area on the transverse plane of the trajectory, third information indicating a degree of difference in the time for support with each of right and left legs in walking, fourth information indicting a degree of difference in the distance of movement while support is made with each of right and left legs in walking, and fifth information indicating a degree of repeating a walking motion during a fixed time.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a walking ability determination device, a walking ability determination method, a walking ability determination program, a model generation device, a model generation method, and a model generation program. [Background technology]

[0002] Various methods for assessing walking ability have been developed. One of the assessment methods is known as FAC (Functional Ambulation Categories) (Non-Patent Document 1). FAC is an assessment method in which a medical professional uses a walkway or staircase of approximately 15 m to assess the walking ability of a subject on a six-point scale based on movement observation. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] "Functional Ambulation Categories (FAC)", [online], [Retrieved May 11, 2021], Internet <URL: http: / / gunma-pt.com / wp-content / uploads / 2018 / 04 / Functional-Ambulation-Categories-FAC.pdf> Summary of the Invention [Problem to be solved by the invention]

[0004] In conventional evaluation methods, experienced medical professionals carefully observe the walking of a subject and evaluate the walking ability of the subject based on the results of the observation. However, training medical professionals to acquire such skills requires time and costs. In addition, because the evaluation is performed by visual observation, there is a problem that the evaluation results may vary depending on the knowledge and experience of the medical professionals.

[0005] In one aspect, the present invention has been made in view of the above circumstances, and an object of the present invention is to provide a technique for appropriately evaluating the walking ability of a subject using a mechanical method. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems, the present invention employs the following configuration.

[0007] That is, a walking ability assessment device according to one aspect of the present invention comprises a data acquisition unit that acquires target data generated by observing the walking of a subject, the target data being configured to include at least one of first information indicating the coronal area of the trajectory of the center of gravity of the subject when walking, second information indicating the area of the trajectory in cross section, third information indicating the degree of difference in the time spent supporting with each leg when walking, fourth information indicating the degree of difference in the distance traveled while supporting with each leg when walking, and fifth information indicating the degree to which walking movements are repeated within a predetermined period of time; a judgment unit that judges the walking ability of the subject based on the acquired target data using a trained judgment model generated by machine learning; and an output unit that outputs the results of the judgment of the walking ability of the subject.

[0008] The present inventors have found through experimental examples described below that it is possible to accurately estimate a subject's walking ability by using, as explanatory variables of a trained model generated by machine learning, at least one of the following: the area in the coronal plane of the trajectory of the center of gravity of the human body during walking, the area in the transverse plane of the trajectory, the degree of difference in the time spent supporting with each leg during walking, the degree of difference in the distance traveled while supporting with each leg during walking, and the degree of repetition of walking movements within a predetermined time. Therefore, with this configuration, it is possible to appropriately evaluate the walking ability of a subject using a mechanical method.

[0009] In the walking ability assessment device according to the above aspect, the target data may be configured to include all of the first information, the second information, the third information, the fourth information, and the fifth information. With this configuration, by using all five pieces of information as explanatory variables, the walking ability of the target person can be estimated with higher accuracy.

[0010] A model generation device according to one aspect of the present invention includes a data acquisition unit that acquires a plurality of learning data sets that are generated by observing the gait of each of a plurality of subjects, each of the learning data sets being configured by a combination of training data and correct answer labels, and the training data of each of the learning data sets includes first information indicating an area in a coronal plane of a trajectory of the center of gravity of each of the subjects when walking, second information indicating an area in a transverse plane of the trajectory, third information indicating a degree of difference in a time period during which the body supports itself with one leg each of the left and right legs when walking, fourth information indicating a degree of difference in a distance traveled while supporting itself with one leg each of the left and right legs when walking, and and a learning processing unit that performs machine learning of a judgment model using the acquired multiple training data sets, wherein the machine learning is configured by training the judgment model so that a result of judging the walking ability of each of the subjects by the judgment model based on the training data for each of the training data sets conforms to the true value indicated by the correct label. With this configuration, it is possible to generate a trained judgment model for appropriately evaluating the walking ability of a subject by a mechanical method.

[0011] In the model generation device according to the above aspect, the training data of each of the learning datasets may be configured to include all of the first information, the second information, the third information, the fourth information, and the fifth information. With this configuration, by using all five pieces of information as explanatory variables, it is possible to generate a trained determination model that can more accurately estimate the walking ability of a subject.

[0012] Furthermore, as another aspect of each of the walking ability assessment device and model generation device according to the above embodiments, one aspect of the present invention may be an information processing method or program that realizes all or part of the above components, or a storage medium that stores such a program and is readable by a computer or other device or machine, or an information processing system. Here, a storage medium that is readable by a computer or the like is a medium that stores information such as a program by electrical, magnetic, optical, mechanical, or chemical action. A walking ability assessment system (information processing system) may be composed of the walking ability assessment device and model generation device according to any of the above embodiments.

[0013] For example, a walking ability assessment method according to one aspect of the present invention is an information processing method that executes the following steps: a step in which a computer acquires target data generated by observing the walking of a subject, the target data being configured to include at least one of first information indicating the area in the coronal plane of the trajectory of the center of gravity of the subject when walking, second information indicating the area in the transverse plane of the trajectory, third information indicating the degree of difference in the time during which the subject supports with each leg when walking, fourth information indicating the degree of difference in the distance traveled while supporting with each leg when walking, and fifth information indicating the degree of repetition of walking movements within a predetermined period of time; a step in which a trained assessment model generated by machine learning is used to assess the walking ability of the subject based on the acquired target data; and a step in which the assessment result of the walking ability of the subject is output.

[0014] For example, a walking ability assessment program according to one aspect of the present invention is a program for causing a computer to execute the steps of: acquiring target data generated by observing the walking of a subject, the target data being configured to include at least one of: first information indicating the area in the coronal plane of the trajectory of the center of gravity of the subject when walking; second information indicating the area in the transverse plane of the trajectory; third information indicating the degree of difference in the time spent supporting with each leg when walking; fourth information indicating the degree of difference in the distance traveled while supporting with each leg when walking; and fifth information indicating the degree of repetition of walking movements within a predetermined period of time; determining the walking ability of the subject based on the acquired target data using a trained assessment model generated by machine learning; and outputting the results of the assessment of the walking ability of the subject.

[0015] For example, a model generation method according to one aspect of the present invention includes a step in which a computer acquires a plurality of learning data sets generated by observing the gait of each of a plurality of subjects, each of the learning data sets being composed of a combination of training data and correct answer labels, and the training data of each of the learning data sets includes first information indicating an area in the coronal plane of a trajectory of the center of gravity of each of the subjects when walking, second information indicating an area in the transverse plane of the trajectory, third information indicating a degree of difference in the time during which the subject supports with one leg each of the left and right legs when walking, and fourth information indicating a degree of difference in the distance traveled while supporting with one leg each of the left and right legs when walking. and a step of performing machine learning of a judgment model using the acquired multiple learning datasets, the machine learning being configured to train the judgment model so that a result of judging the walking ability of each of the subjects by the judgment model based on the training data for each of the learning datasets matches the true value indicated by the correct label.

[0016] For example, a model generation program according to one aspect of the present invention includes a step of acquiring, in a computer, a plurality of learning data sets that have been generated by observing the gait of each of a plurality of subjects, each of the learning data sets being configured by a combination of training data and correct answer labels, and the training data of each of the learning data sets includes first information indicating the coronal area of the trajectory of the center of gravity of each of the subjects when walking, second information indicating the area of the trajectory in a cross section, third information indicating the degree of difference in the time during which the subject supports the body with one leg each on the left and right during walking, and fourth information indicating the degree of difference in the distance traveled while supporting the body with one leg each on the left and right during walking. and fifth information indicating the degree to which the walking action is repeated within a predetermined period of time, and the correct answer label of each of the learning datasets is configured to indicate a true value of the walking ability of each of the subjects; and a step of performing machine learning of a judgment model using the acquired multiple learning datasets, the machine learning being configured to train the judgment model so that, for each of the learning datasets, the results of judging the walking ability of each of the subjects by the judgment model based on the training data conform to the true value indicated by the correct answer label. [Effects of the Invention]

[0017] According to the present invention, it is possible to provide a technique for appropriately evaluating the walking ability of a subject using a mechanical method. [Brief explanation of the drawings]

[0018] [Figure 1] FIG. 1 shows a schematic diagram of an example of a situation in which the present invention is applied. [Figure 2] FIG. 2 is a diagram illustrating an example of a hardware configuration of a model generating device according to an embodiment. [Figure 3] FIG. 3 is a schematic diagram showing an example of the hardware configuration of the walking ability determination device according to the embodiment. [Figure 4]FIG. 4 is a diagram illustrating an example of a software configuration of the model generating device according to the embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of the software configuration of the walking ability determination device according to the embodiment. [Figure 6] FIG. 6 is a flowchart illustrating an example of a processing procedure of the model generating device according to the embodiment. [Figure 7] FIG. 7 shows the results of measuring the walking of a subject using an acceleration sensor. [Figure 8A] FIG. 8A shows the trajectory (coronal plane) of the center of gravity of a person while walking, obtained by analyzing the measurement results of FIG. [Figure 8B] FIG. 8B shows the trajectory (coronal plane) of the center of gravity of a person while walking, obtained by applying filtering processing to the calculation results of FIG. 8A. [Figure 9A] FIG. 9A shows the trajectory (cross section) of the center of gravity of a person while walking, obtained by analyzing the measurement results of FIG. [Figure 9B] FIG. 9B shows the trajectory (cross section) of the center of gravity of a person while walking, obtained by applying filtering processing to the calculation results of FIG. 9A. [Figure 10] FIG. 10 is a flowchart illustrating an example of a processing procedure of the walking ability determination device according to the embodiment. [Figure 11] FIG. 11 shows the results of calculating the contribution rate of each piece of information. DETAILED DESCRIPTION OF THE INVENTION

[0019] An embodiment according to one aspect of the present invention (hereinafter also referred to as "the present embodiment") will be described below with reference to the drawings. However, the present embodiment described below is merely an example of the present invention in all respects. Needless to say, various improvements and modifications can be made without departing from the scope of the present invention. In implementing the present invention, a specific configuration according to the embodiment may be adopted as appropriate. Note that while data appearing in this embodiment is described in natural language, more specifically, it is specified using computer-recognizable pseudo-language, commands, parameters, machine language, etc.

[0020] §1 Application Examples 1 is a schematic diagram illustrating an example of a situation in which the present invention is applied. As shown in FIG. 1, a determination system 100 according to this embodiment includes a model generation device 1 and a walking ability determination device 2.

[0021] The model generation device 1 according to this embodiment is a computer configured to generate a trained determination model 6 that can be used to evaluate the walking ability of a subject. The model generation device 1 acquires multiple learning datasets 3 that are generated by observing the gait of each of multiple subjects.

[0022] Each learning data set 3 is composed of a combination of training data 31 and correct labels 32. The training data 31 of each learning data set 3 is configured to include at least one of: first information indicating the area in the coronal plane of the trajectory of the center of gravity of each subject when walking; second information indicating the area in the transverse plane of the trajectory; third information indicating the degree of difference in the time spent supporting on each leg when walking; fourth information indicating the degree of difference in the distance traveled while supporting on each leg when walking; and fifth information indicating the degree of repetition of walking movements within a predetermined time. The correct labels 32 of each learning data set 3 are configured to indicate the true value of each subject's walking ability.

[0023] The model generation device 1 performs machine learning of the determination model 6 using the acquired multiple training data sets 3. The machine learning is configured by training the determination model 6 so that the result of determining the walking ability of each subject by the determination model 6 based on the training data 31 for each training data set 3 matches the true value indicated by the correct label 32. This makes it possible to generate a trained determination model 6 that has acquired the ability to estimate the walking ability of a subject based on at least any of the first to fifth information.

[0024] On the other hand, the walking ability assessment device 2 is a computer configured to assess the walking ability of a subject using a trained assessment model 6. The walking ability assessment device 2 acquires subject data 221 generated by observing the walking of the subject. The subject is a person whose walking ability is to be assessed. The subject may be any one of the multiple subjects during the machine learning, or may be a person other than the multiple subjects.

[0025] The target data 221 is configured to include at least any of the following: first information indicating the area in the coronal plane of the trajectory of the center of gravity of the subject when walking; second information indicating the area in the transverse plane of the trajectory; third information indicating the degree of difference in the time during which the subject supports with each leg when walking; fourth information indicating the degree of difference in the distance traveled while supporting with each leg when walking; and fifth information indicating the degree of repetition of walking movements within a predetermined time. The information included in the target data 221 is configured to correspond to the information included in the training data 31.

[0026] The walking ability determination device 2 uses the trained determination model 6 generated by the above-mentioned machine learning to determine the walking ability of the subject based on the acquired subject data 221. Then, the walking ability determination device 2 outputs the determined walking ability of the subject.

[0027] As described above, in this embodiment, at least one of the first information to the fifth information is adopted as an explanatory variable for estimating walking ability, thereby making it possible to appropriately evaluate the walking ability of a subject using a mechanical method.

[0028] In the example of Fig. 1, the model generating device 1 and the walking ability determination device 2 are connected to each other via a network. The type of network may be appropriately selected from, for example, the Internet, a wireless communication network, a mobile communication network, a telephone network, a dedicated network, etc. However, the method of exchanging data between the model generating device 1 and the walking ability determination device 2 is not limited to this example and may be appropriately selected depending on the embodiment. As another example, data may be exchanged between the model generating device 1 and the walking ability determination device 2 using a storage medium.

[0029] 1, the model generation device 1 and the walking ability determination device 2 are each configured as separate computers. However, the configuration of the determination system 100 according to this embodiment is not limited to this example and may be determined appropriately depending on the embodiment. As another example, the model generation device 1 and the walking ability determination device 2 may be integrated into one computer. As yet another example, at least one of the model generation device 1 and the walking ability determination device 2 may be configured as multiple computers.

[0030] Any sensor may be used to observe the gait of the subject / object. The type of sensor is not particularly limited as long as it can analyze the information used for the explanatory variables among the first information to the fifth information, and may be appropriately selected depending on the embodiment. The sensor may be, for example, an acceleration sensor, a motion capture sensor, a camera (e.g., multiple cameras, a depth camera, a stereo camera, etc.), etc. The sensor may be appropriately installed so that it can analyze the information used for the explanatory variables. As an example, when an acceleration sensor is used as the sensor, the acceleration sensor may be attached to the waist of the subject / object.

[0031] §2 Configuration example [Hardware configuration] <Model generation device> Fig. 2 schematically illustrates an example of the hardware configuration of a model generation device 1 according to this embodiment. As shown in Fig. 2, the model generation device 1 according to this embodiment is a computer to which a control unit 11, a storage unit 12, a communication interface 13, an external interface 14, an input device 15, an output device 16, and a drive 17 are electrically connected. In Fig. 2, the communication interface and the external interface are referred to as a "communication I / F" and an "external I / F." Similar notations are used in Fig. 3, which will be described later.

[0032] The control unit 11 includes a hardware processor such as a central processing unit (CPU), a random access memory (RAM), and a read-only memory (ROM), and is configured to execute information processing based on programs and various data. The storage unit 12 is an example of memory, and is configured, for example, with a hard disk drive or a solid-state drive. In this embodiment, the storage unit 12 stores various information such as a model generation program 81, multiple training datasets 3, and training result data 125.

[0033] The model generation program 81 is a program for causing the model generation device 1 to execute information processing ( FIG. 6 ) described below related to the generation of a trained judgment model 6 (machine learning of the judgment model 6). The model generation program 81 includes a series of instructions for the information processing. A plurality of learning datasets 3 are used to generate the trained judgment model 6. The learning result data 125 indicates information related to the trained judgment model 6. In this embodiment, the learning result data 125 is generated as a result of executing the model generation program 81. Details will be described later.

[0034] The communication interface 13 is, for example, a wired LAN (Local Area Network) module, a wireless LAN module, etc., and is an interface for performing wired or wireless communication via a network. The model generation device 1 can use the communication interface 13 to perform data communication with other information processing devices via the network.

[0035] The external interface 14 is, for example, a USB (Universal Serial Bus) port, a dedicated port, or the like, and is an interface for connecting to an external device. The type and number of external interfaces 14 may be selected arbitrarily. When a sensor is used to generate each training dataset 3 in the model generation device 1, the sensor may be connected via the communication interface 13 or the external interface 14.

[0036] The input device 15 is a device for inputting, for example, a mouse, a keyboard, etc. The output device 16 is a device for outputting, for example, a display, a speaker, etc. An operator such as a user can operate the model generation device 1 by using the input device 15 and the output device 16.

[0037] The drive 17 is, for example, a CD drive, a DVD drive, or the like, and is a drive device for reading various information, such as programs, stored in a storage medium 91. The storage medium 91 is a medium that stores information, such as programs, by electrical, magnetic, optical, mechanical, or chemical action so that a computer or other device, machine, or the like can read the stored information. At least one of the model generation program 81 and the training dataset 3 may be stored in the storage medium 91. The model generation device 1 may acquire at least one of the model generation program 81 and the training dataset 3 from the storage medium 91. Note that FIG. 2 illustrates a disk-type storage medium, such as a CD or a DVD, as an example of the storage medium 91. However, the type of the storage medium 91 is not limited to a disk-type storage medium and may be other than a disk-type storage medium. Examples of storage media other than a disk-type storage medium include semiconductor memories, such as flash memories. The type of the drive 17 may be selected arbitrarily depending on the type of the storage medium 91.

[0038] Note that, with regard to the specific hardware configuration of the model generation device 1, components can be omitted, replaced, or added as appropriate depending on the embodiment. For example, the control unit 11 may include multiple hardware processors. The hardware processor may be configured with a microprocessor, an FPGA (field-programmable gate array), or the like. The storage unit 12 may be configured with RAM and ROM included in the control unit 11. At least one of the communication interface 13, the external interface 14, the input device 15, the output device 16, and the drive 17 may be omitted. The model generation device 1 may be configured with multiple computers. In this case, the hardware configurations of the computers may or may not be the same. Furthermore, the model generation device 1 may be an information processing device designed specifically for the service to be provided, as well as a general-purpose server device, a general-purpose PC (Personal Computer), or the like.

[0039] <Walking ability assessment device> Fig. 3 schematically illustrates an example of the hardware configuration of the walking ability assessment device 2 according to this embodiment. As shown in Fig. 3, the walking ability assessment device 2 according to this embodiment is a computer to which a control unit 21, a storage unit 22, a communication interface 23, an external interface 24, an input device 25, an output device 26, and a drive 27 are electrically connected.

[0040] The control unit 21 to the drive 27 and the storage medium 92 of the walking ability determination device 2 may be configured similarly to the control unit 11 to the drive 17 and the storage medium 91 of the model generation device 1, respectively. The control unit 21 includes a hardware processor such as a CPU, RAM, and ROM, and is configured to execute various information processes based on programs and data. The storage unit 22 is configured, for example, with a hard disk drive or a solid state drive. In this embodiment, the storage unit 22 stores various information such as a walking ability determination program 82 and learning result data 125.

[0041] The walking ability determination program 82 is a program for causing the walking ability determination device 2 to execute information processing (FIG. 10) described below related to determining the walking ability of the subject. The walking ability determination program 82 includes a series of instructions for this information processing. Details will be described later. At least one of the walking ability determination program 82 and the learning result data 125 may be stored in the storage medium 92. Furthermore, the walking ability determination device 2 may acquire at least one of the walking ability determination program 82 and the learning result data 125 from the storage medium 92.

[0042] Note that, with regard to the specific hardware configuration of the walking ability assessment device 2, components can be omitted, replaced, or added as appropriate depending on the embodiment. For example, the control unit 21 may include multiple hardware processors. The hardware processor may be configured with a microprocessor, FPGA, etc. The storage unit 22 may be configured with RAM and ROM included in the control unit 21. At least one of the communication interface 23, the external interface 24, the input device 25, the output device 26, and the drive 27 may be omitted. The walking ability assessment device 2 may be configured with multiple computers. In this case, the hardware configurations of the computers may or may not be the same. Furthermore, the walking ability assessment device 2 may be an information processing device designed specifically for the services provided, as well as a general-purpose server device, a general-purpose PC, a mobile terminal (e.g., a smartphone), a tablet PC, a microcomputer, etc.

[0043] [Software configuration] <Model generation device> 4 schematically illustrates an example of the software configuration of the model generation device 1 according to this embodiment. The control unit 11 of the model generation device 1 loads a model generation program 81 stored in the storage unit 12 into RAM. The control unit 11 then executes instructions included in the model generation program 81 loaded into RAM using the CPU to control each component. As a result, as shown in FIG. 4, the model generation device 1 according to this embodiment operates as a computer including a training data acquisition unit 111, a learning processing unit 112, and a storage processing unit 113 as software modules.

[0044] The learning data acquisition unit 111 is configured to acquire a plurality of learning data sets 3 generated by observing the gait of each of a plurality of subjects. Each learning data set 3 is composed of a combination of training data 31 and a correct answer label 32. The training data 31 is configured to include at least any of the above-mentioned first information to fifth information of the corresponding subject. The correct answer label 32 is configured to indicate the true value of the walking ability of the corresponding subject.

[0045] The learning processing unit 112 is configured to use the acquired multiple learning datasets 3 to perform machine learning of the determination model 6. The machine learning is configured by training the determination model 6 so that, for each learning dataset 3, the result of determining the walking ability of each subject by the determination model 6 based on the training data 31 matches the true value indicated by the correct label 32.

[0046] The saving processing unit 113 is configured to generate information about the trained judgment model 6 generated by machine learning as learning result data 125, and to save the generated learning result data 125 in an arbitrary storage area. The learning result data 125 may be appropriately configured to include information for reproducing the trained judgment model 6.

[0047] (An example of a decision model) The determination model 6 is configured by a machine learning model having one or more calculation parameters used in a calculation to derive an inference result. The type of machine learning model is not particularly limited as long as it can execute the process of inferring walking ability, and may be appropriately selected depending on the embodiment. As an example, the determination model 6 may be configured by a neural network, a support vector machine, a regression model, a decision tree model, etc.

[0048] Training the decision model 6 involves adjusting (optimizing) the values of calculation parameters using each learning dataset 3 so as to derive the true value of the corresponding correct label 32 from the training data 31. The machine learning method may be selected appropriately depending on the type of machine learning model employed. For example, the machine learning method may be a backpropagation method, a method that solves an optimization problem, a method that performs regression analysis, a random forest, or the like.

[0049] In the example of FIG. 4, the determination model 6 is configured by a neural network. When a neural network is used to configure the determination model 6, the determination model 6 is typically configured to include an input layer, one or more intermediate layers (hidden layers), and an output layer. Any type of layer, such as a fully connected layer, may be used for each layer. The number of layers included in the determination model 6, the number of nodes (neurons) in each layer, and the connection relationships between the nodes may be determined appropriately depending on the embodiment. The weight of the connections between each node, the threshold value of each node, etc. are examples of the above-mentioned calculation parameters.

[0050] As an example of a training process when a neural network is employed, the learning processing unit 112 inputs the training data 31 of each learning data set 3 into the determination model 6 and executes a forward propagation calculation process for the determination model 6. As a calculation result of this forward propagation, the learning processing unit 112 acquires an output value corresponding to the result of judging the walking ability of each subject based on the training data 31 of each learning data set 3. The learning processing unit 112 calculates the error between the acquired output value and the true value indicated by the corresponding correct label 32, and further calculates the gradient of the calculated error. Next, the learning processing unit 112 backpropagates the calculated gradient of the error using an error backpropagation method to calculate the error of the value of the calculation parameter of the determination model 6. Then, the learning processing unit 112 updates the value of the calculation parameter based on the calculated error.

[0051] Through this series of update processes, the learning processing unit 112 adjusts the parameter values of the determination model 6 so as to reduce the sum of errors between the determination result (output value) and the true value for each learning dataset 3. This adjustment of the parameter values may be repeated until a predetermined condition is met, such as a specified number of times or the calculated sum of errors becoming equal to or less than a threshold. Furthermore, machine learning conditions such as a loss function and a learning rate may be set appropriately depending on the embodiment. Through this machine learning process, a trained determination model 6 that has acquired the ability to determine the walking ability of a subject based on at least one of the first to fifth information can be generated.

[0052] The storage processing unit 113 stores the trained judgment model 6 generated by the above machine learning as learning result data 125. The configuration of the learning result data 125 is not particularly limited as long as it can hold information for executing calculations of the trained judgment model 6, and may be determined appropriately depending on the embodiment. As an example, the learning result data 125 may be configured to include information indicating the configuration of the judgment model 6 (e.g., the structure of a neural network) and the values of calculation parameters adjusted by machine learning. The learning result data 125 may be stored in any storage area. The learning result data 125 may be referenced as appropriate to set the trained judgment model 6 in a usable state on a computer.

[0053] <Walking ability assessment device> 5 schematically illustrates an example of the software configuration of the walking ability assessment device 2 according to this embodiment. The control unit 21 of the walking ability assessment device 2 loads a walking ability assessment program 82 stored in the storage unit 22 into RAM. The control unit 21 then causes the CPU to execute instructions included in the walking ability assessment program 82 loaded into RAM, thereby controlling each component. As a result, as shown in FIG. 5, the walking ability assessment device 2 according to this embodiment operates as a computer including a data acquisition unit 211, a determination unit 212, and an output unit 213 as software modules.

[0054] The data acquisition unit 211 is configured to acquire target data 221 generated by observing the walking of the target person. The target data 221 is configured to include at least any of the first information to the fifth information of the target person. The determination unit 212 holds the learning result data 125, and is thus provided with a trained determination model 6 generated by machine learning. The determination unit 212 is configured to determine the walking ability of the target person based on the acquired target data 221, using the trained determination model 6. The output unit 213 is configured to output the determination result of the walking ability of the target person.

[0055] <Other> Each software module of the model generating device 1 and the walking ability assessment device 2 will be described in detail in the operation example below. Note that in this embodiment, an example is described in which each software module of the model generating device 1 and the walking ability assessment device 2 is realized by a general-purpose CPU. However, some or all of the above software modules may be realized by one or more dedicated processors. In other words, each of the above modules may be realized as a hardware module. Furthermore, with regard to the software configuration of each of the model generating device 1 and the walking ability assessment device 2, software modules may be omitted, replaced, or added as appropriate depending on the embodiment.

[0056] §3 Example of operation [Model generation device] FIG. 6 is a flowchart showing an example of a processing procedure for machine learning of a determination model 6 by the model generation device 1 according to this embodiment. The following processing procedure of the model generation device 1 is an example of a model generation method. However, the following processing procedure of the model generation device 1 is merely an example, and each step may be changed as much as possible. Furthermore, steps may be omitted, replaced, or added to the following processing procedure of the model generation device 1 as appropriate depending on the embodiment.

[0057] (Step S101) In step S101, the control unit 11 operates as the learning data acquisition unit 111 and acquires a plurality of learning data sets 3.

[0058] Each learning data set 3 may be generated as needed by observing the gait of the subject. Any sensor may be used to observe the gait of the subject. The sensor may be, for example, an acceleration sensor, a motion capture device, a camera, or the like. By appropriately analyzing the data obtained by the sensor, at least one of the first information to the fifth information can be obtained, and the training data 31 can be generated from this.

[0059] The first and second information can be obtained by three-dimensionally analyzing the movement of the center of gravity of a person from data obtained by a sensor. Furthermore, the starting point of a walking movement can be set at any timing, and one walking movement can be captured from the start of walking to the time when the person returns to the starting point. As an example, assuming that the starting point of a walking movement is the time when one leg touches the ground, one walking movement corresponds to a series of movements including a period in which the body is supported by both legs after one leg touches the ground, a period in which the other leg lifts off the ground and the body is supported only by one leg, a period in which the other leg touches the ground and the body is supported only by both legs, and a period in which one leg lifts off the ground and the body is supported only by the other leg, until one leg touches the ground again. The third to fifth information can be obtained by analyzing such walking movements from data obtained by a sensor.

[0060] FIG. 7 shows the results (measurement data) of measuring the walking of a subject using an acceleration sensor in an experimental example described below. FIG. 8A shows the trajectory (coronal plane) of the center of gravity of the person while walking, obtained by analyzing the measurement results of FIG. 7. FIG. 8B shows the trajectory (coronal plane) of the center of gravity of the person while walking, obtained by applying filtering processing to the calculation results of FIG. 8A. FIG. 9A shows the trajectory (cross section) of the center of gravity of the person while walking, obtained by analyzing the measurement results of FIG. 7. FIG. 9B shows the trajectory (cross section) of the center of gravity of the person while walking, obtained by applying filtering processing to the calculation results of FIG. 9A.

[0061] In this gait measurement, the y-axis of the three-dimensional space corresponds to the front-to-back direction of the subject (the forward direction is the walking direction), the x-axis corresponds to the left-to-right direction, and the z-axis corresponds to the up-to-down direction. However, the correspondence between the directions is not limited to this example. The correspondence between the directions may be determined appropriately depending on the embodiment.

[0062] As an example, the coronal plane trajectory shown in FIG. 8A can be obtained by integrating the measurement data of the acceleration sensor shown in FIG. 7 twice and plotting the left-right position and the up-down position at each time as a two-dimensional vector. Filtering (Fourier transform, then band-pass filtering for each of several peaks, followed by inverse Fourier transform) of this trajectory can be applied to obtain the trajectory shown in FIG. 8B. The first information (D_xz) can be obtained by calculating the area of this trajectory through integration. Meanwhile, the transverse plane trajectory shown in FIG. 9A can be obtained by integrating the measurement data shown in FIG. 7 twice and plotting the left-right position and the up-down position at each time as a two-dimensional vector. Similarly to the transverse plane trajectory, filtering can be applied to this trajectory to obtain the trajectory shown in FIG. 9B. The second information (D_xy) can be obtained by calculating the area of this trajectory through integration.

[0063] Furthermore, by converting the acceleration sensor measurement data as shown in Figure 8B and measuring the time (t1) of the right vertical minimum point and the time (t2) of the left vertical minimum point of the infinity (∞)-shaped plot, the time when each foot switches between swing and stance can be determined, allowing the time each leg supports on one leg during walking to be calculated. The calculated support time for each leg can be obtained from the third information (γ_t). Furthermore, by converting the acceleration sensor measurement data as shown in Figures 8B and 9B and treating the vertical (z-axis), horizontal (x-axis), and front-to-back (y-axis) positions at each time as three-dimensional vectors, the trajectory of the hips can be reconstructed in three dimensions. The boundaries between the swing and stance legs of the left and right feet on the trajectory are indicated by the vertical minimum points on the left and right as described above, and by measuring the distance along the three-dimensional trajectory between them for the time of the left stance leg and the time of the right stance leg, respectively, the distance traveled while supporting on one leg during walking can be calculated. Strictly speaking, there is a period of time when both feet are in stance, but this period is considered to be infinitesimally short, and the left and right stance legs are considered to switch positions at the minimum point. The fourth information (γ_d) can be obtained from the calculated distance traveled on each leg. As long as it is possible to express the degree of left-right imbalance, the "degree of difference" in the third information and the fourth information may be expressed in any manner. As an example, the "degree of difference" in the third information and the fourth information may be expressed as a left-right ratio or a left-right difference.

[0064] Furthermore, when the acceleration sensor measurement data is Fourier transformed, several peaks appear. The lowest frequency peak indicates the base frequency of the person's walking, which is the stride frequency (unit: Hz) (two steps, one on each side). Its reciprocal is the time for two steps (unit: seconds). Furthermore, in Figure 8B or Figure 9B, this two-step time corresponds to the time it takes to go around from a starting point to the same point and return to that starting point. Halving this equals the average time for one step. The peaks other than the base frequency are its harmonics, allowing the repetition of walking movements to be captured as a whole. By counting the number of repetitions of walking movements within a specified time, the fifth information can be obtained using this base frequency. The specified time may be determined as appropriate. Note that the "degree of repetition of walking movements" in the fifth information may be expressed in any manner, as long as it can represent whether the number of steps is high or low. As an example, the "degree of repetition of walking movements" may be expressed using frequency or period.

[0065] The training data 31 is configured to include at least any of the first to fifth information. The training data 31 of each learning dataset 3 may be configured to include all of the first, second, third, fourth, and fifth information of each subject. Furthermore, information other than these may be further adopted as explanatory variables for inference. When other information is further adopted as explanatory variables, the training data 31 may be configured to further include the other information.

[0066] Meanwhile, the correct label 32 can be generated by appropriately evaluating the walking ability of each subject. The evaluation of the walking ability of each subject may be performed manually (by a doctor). Alternatively, the evaluation of the walking ability of at least some of the subjects may be performed by any mechanical method. The correct label 32 is configured to indicate the true value of the walking ability of the subject. The evaluation index for walking ability may be any. Walking ability may be expressed by a continuous value or a discrete value (e.g., a category, a class, etc.). As an example, FAC may be adopted as the evaluation index for walking ability. Each learning dataset 3 can be generated by associating the generated correct label 32 with the corresponding training data 31. The number of learning datasets 3 generated from one subject may be any.

[0067] Each training dataset 3 may be generated automatically by the operation of a computer, or may be generated manually with at least partial operation by an operator. Furthermore, each training dataset 3 may be generated by the model generation device 1, or by a computer other than the model generation device 1. When each training dataset 3 is generated by the model generation device 1, the control unit 11 acquires each training dataset 3 by automatically executing the generation process or manually by an operator. On the other hand, when each training dataset 3 is generated by one or more other computers, the control unit 11 acquires each training dataset 3, for example, via a network, a storage medium 91, an external storage device, or the like. Some training datasets 3 may be generated by the model generation device 1, and the other training datasets 3 may be generated by one or more other computers.

[0068] The number of acquired training data sets 3 is not particularly limited and may be determined appropriately depending on the embodiment. After acquiring a plurality of training data sets 3, the control unit 11 proceeds to the next step S102.

[0069] (Step S102) In step S102, the control unit 11 operates as the learning processing unit 112 and performs machine learning of the judgment model 6 using the acquired multiple learning datasets 3. As described above, the control unit 11 optimizes the values of the calculation parameters of the judgment model 6 through machine learning so as to reduce the error between the inference result (the result of determining walking ability) for the training data 31 of each learning dataset 3 and the true value indicated by the correct label 32. As a result of this machine learning, it is possible to generate a trained judgment model 6 that has acquired the ability to determine the walking ability of a subject based on at least any of the first information to the fifth information. If the training data 31 includes all of the first information to the fifth information, it is possible to generate a trained judgment model 6 that has acquired the ability to determine the walking ability of a subject based on all of the first information to the fifth information. When the machine learning process is completed, the control unit 11 proceeds to the next step S103.

[0070] (Step S103) In step S103, the control unit 11 operates as the storage processing unit 113 and generates information about the trained determination model 6 generated by machine learning as learning result data 125. Then, the control unit 11 stores the generated learning result data 125 in an arbitrary storage area.

[0071] The learning result data 125 may be stored in, for example, RAM in the control unit 11, the storage unit 12, an external storage device, a storage medium, or a combination of these. The storage medium may be, for example, a CD or a DVD, and the control unit 11 may store the learning result data 125 in the storage medium via the drive 17. The external storage device may be, for example, a data server such as a network-attached storage (NAS). In this case, the control unit 11 may use the communication interface 13 to store the learning result data 125 in the data server via a network. The external storage device may also be, for example, an external storage device connected to the model generation device 1 via the external interface 14.

[0072] When the storage of the learning result data 125 is completed, the control unit 11 ends the processing procedure of the model generating device 1 according to this operation example.

[0073] The generated learning result data 125 may be provided to the walking ability determination device 2 at any timing. As an example, the control unit 11 may transfer the learning result data 125 to the walking ability determination device 2 as part of the processing of step S103 above or separately from the processing of step S103. The walking ability determination device 2 may acquire the learning result data 125 by receiving this transfer. As another example, the walking ability determination device 2 may acquire the learning result data 125 by accessing the model generation device 1 or a data server via a network using the communication interface 23. As another example, the walking ability determination device 2 may acquire the learning result data 125 via the storage medium 92. As another example, the learning result data 125 may be pre-installed in the walking ability determination device 2.

[0074] Furthermore, the control unit 11 may update or newly generate the trained judgment model 6 (learning result data 125) by periodically or irregularly repeating the processes from step S101 to step S103. During this repetition, at least a portion of the learning dataset 3 used for machine learning may be changed, modified, added, deleted, or the like as appropriate. Then, the control unit 11 may update the learning result data 125 held by the walking ability judgment device 2 by providing the updated or newly generated learning result data 125 to the walking ability judgment device 2 by any method and at any timing.

[0075] [Walking ability assessment device] 10 is a flowchart showing an example of a processing procedure for determining the walking ability of a subject by the walking ability determination device 2 according to this embodiment. The processing procedure of the walking ability determination device 2 described below is an example of a walking ability determination method. However, the processing procedure of the walking ability determination device 2 described below is merely an example, and each step may be modified as much as possible. Furthermore, steps may be omitted, replaced, or added to the following processing procedure as appropriate depending on the embodiment.

[0076] (Step S201) In step S201, the control unit 21 operates as the data acquisition unit 211 and acquires the target data 221.

[0077] The target data 221 is the same type of data as the training data 31, and may be generated in the same manner as the training data 31. The target data 221 is configured to include at least any of the first information to the fifth information of the subject. The target data 221 may be configured to include all of the first information, second information, third information, fourth information, and fifth information of the subject.

[0078] In one example, the control unit 21 may acquire measurement data generated by observing the walking of the subject from a sensor directly or indirectly, and analyze the acquired measurement data to acquire the target data 221. In another example, the target data 221 may be generated by another computer. The control unit 21 may acquire the target data 221, for example, via a network, a storage medium 92, an external storage device, etc. After acquiring the target data 221, the control unit 21 proceeds to the next step S202.

[0079] (Step S202) In step S202, the control unit 21 operates as the judgment unit 212 and sets the trained judgment model 6 with reference to the learning result data 125. Then, the control unit 21 uses the trained judgment model 6 to judge the walking ability of the subject based on the acquired target data 221. That is, the control unit 21 inputs the acquired target data 221 to the trained judgment model 6 and executes arithmetic processing of the trained judgment model 6. As a result of executing this arithmetic processing, the control unit 21 acquires an output value corresponding to the result of judging the walking ability of the subject from the trained judgment model 6. Upon acquiring the judgment result, the control unit 21 proceeds to the next step S203.

[0080] (Step S203) In step S203, the control unit 21 operates as the output unit 213 and outputs the assessment result of the walking ability of the subject. The output destination of the assessment result and the content of the information to be output may each be appropriately determined depending on the embodiment. For example, the control unit 21 may output the assessment result obtained in step S202 as is. Alternatively, the control unit 21 may perform some information processing based on the assessment result obtained. Then, the control unit 21 may output the result of the information processing as information related to the assessment result. As an example, a correspondence relationship between the assessment result of the walking ability and information to be notified to the subject (e.g., rehabilitation details for a subject with low walking ability) may be determined in advance. Based on this correspondence relationship, the control unit 21 may generate notification information for the subject from the assessment result obtained in step S202 and output the generated notification information. The output destination may be, for example, the output device 26, an output device of another computer, etc. The output format may be, for example, screen output, audio output, printout, etc.

[0081] When the output of information related to the determination result is completed, the control unit 21 ends the processing procedure of the walking ability determination device 2 according to this operation example. The control unit 21 may acquire target data 221 of multiple subjects in step S201, and execute the processing of steps S202 and S203 for the target data 221 of each subject. In this way, the walking ability determination device 2 may determine the walking ability of each of the multiple subjects. Furthermore, the control unit 21 may repeatedly execute the series of information processing from step S201 to step S203 for any subject. In this way, the walking ability determination device 2 may continuously monitor the walking ability of the subject.

[0082] [Features] As described above, in this embodiment, the processing of steps S101 to S103 can generate a trained determination model 6 that has acquired the ability to determine walking ability from at least any of the first information to the fifth information. In the processing of step S202, by using the generated trained determination model 6, the walking ability of the subject can be appropriately evaluated by a mechanical method. Furthermore, by adopting all of the first information to the fifth information as explanatory variables of the determination model 6, the walking ability of the subject can be determined more accurately.

[0083] §4 Experimental Examples <First Experimental Example> Each of the 27 subjects was fitted with an acceleration sensor (IMU-Z, manufactured by ZMP) on their waist, and the gait of each subject was monitored using the acceleration sensor to obtain measurement data (Figure 7). Five pieces of information (gait data), numbered 1 through 5, were calculated from the obtained measurement data using the method described above. Differences were used to represent the third and fourth pieces of information. Frequency was used to represent the fifth piece of information (freq). Additionally, 19 medical professionals (doctors) evaluated each subject's walking ability using the FAC index, and the average of the medical professionals' evaluation results was obtained as the true value of each subject's walking ability. The obtained true values were then associated with the walking data as correct labels to generate a dataset.

[0084] Of the generated datasets for 29 subjects, datasets for 15 subjects were used for machine learning to generate a trained judgment model. A layered neural network (number of layers: 3, number of nodes: 5, 3, 1) was used to configure the judgment model, and the above-mentioned backpropagation method was used as the machine learning method. The remaining datasets for 12 subjects were then used to evaluate the generated trained judgment model. The evaluation results of the walking ability of each subject are shown in Table 1 below.

[0085] [Table 1]

[0086] As shown in Table 1, the average error between the judgment results of the generated trained judgment model and the average (true value) of the walking ability assessment results by medical professionals was 0.996. Considering that the dataset used for machine learning was for 15 people, it was found that the walking ability of the subjects could be judged with a fair degree of accuracy based on the above five pieces of information.

[0087] Furthermore, when looking at the assessment results for each subject, there was a somewhat large discrepancy in the assessment of subjects around FAC3 and 4. This was presumably due to the fact that even experienced physicians are prone to making mistakes when assessing walking ability around FAC3 and 4 (for example, mistaking FAC3 for FAC4), which resulted in deviations in the true values indicated by the correct labels. Therefore, based on these points, it was presumed that a more accurate trained assessment model could be generated by increasing the number of datasets used for learning and the number of physicians who evaluate each subject's walking ability when obtaining the true value (correct label) of walking ability.

[0088] <Second Experimental Example> Next, a pair of subjects was randomly selected using a random number, and one of the five pieces of information was exchanged (for example, the value of the fifth piece of information was exchanged between Patient A and Patient B). The difference in the output value of the trained judgment model generated in the first experimental example before and after the exchange was calculated as the permutation importance (PI) value of each piece of information. The ratio of each PI value to the sum of the absolute values of the calculated PI values of the five pieces of information was then calculated as the contribution rate of each piece of information to the assessment of walking ability.

[0089] FIG. 11 shows the results of calculating the contribution rate of each piece of information. As shown in FIG. 11, each of the five pieces of information had a certain degree of contribution rate. From these results, it was found that it is possible to determine the walking ability of a subject without using all five pieces of information (i.e., by using at least one of the five pieces of information). Furthermore, among the five pieces of information, the contribution rates of the first piece of information (the area in the coronal plane of the trajectory of the center of gravity of the human body) and the fourth piece of information (the difference between the left and right movement distances when supporting on one leg) were relatively high. From these results, it was found that it is preferable that the training data 31 and the subject data 221 be configured to include at least one of the first piece of information and the fourth piece of information. [Explanation of symbols]

[0090] 1...Model generation device, 11...control unit, 12...storage unit, 13...communication interface, 14...External interface, 15...input device, 16...output device, 17...drive, 81...model generation program, 91...storage medium, 111... learning data acquisition unit, 112... learning processing unit, 113...storage processing unit, 125...learning result data, 2... Walking ability assessment device, 21...control unit, 22...storage unit, 23...communication interface, 24...External interface, 25...input device, 26...output device, 27...drive, 82...walking ability determination program, 92...storage medium, 211...data acquisition unit, 212...determination unit, 213...output unit, 221...Target data, 3...training dataset, 31...Training data, 32...Correct label, 6...Decision model

Claims

1. a data acquisition unit configured to acquire target data generated by observing the walking of a target person, the target data including at least one of first information indicating an area in the coronal plane of a locus of the center of gravity of the target person when walking, second information indicating an area in the transverse plane of the locus, third information indicating a degree of difference in the time during which the target person supports with each of the left and right legs when walking, fourth information indicating a degree of difference in the distance traveled while supporting with each of the left and right legs when walking, and fifth information indicating a degree of repetition of a walking action within a predetermined time; a determination unit that determines the walking ability of the subject based on the acquired subject data using a trained determination model generated by machine learning; an output unit that outputs the result of determining the walking ability of the subject; Equipped with the target data is configured to include all of the first information, the second information, the third information, the fourth information, and the fifth information, the target data is measurement data of an acceleration sensor attached to the waist of the target person, the first information is obtained by integrating the measurement data twice and plotting the left-right position and the up-down position at each time as a two-dimensional vector to obtain a coronal plane trajectory, and applying a filtering process to the coronal plane trajectory to calculate an area of the trajectory by integration; the second information is obtained by integrating the measurement data twice and plotting the left and right positions and the front and rear positions at each time as two-dimensional vectors to obtain a trajectory of the cross section, and applying a filtering process to the trajectory to calculate an area of the trajectory by integration; the third information is obtained from a time period during which the patient supports the body on one leg, the time period being calculated by measuring a time of a minimum point in the right vertical direction and a time of a minimum point in the left vertical direction of a trajectory obtained by applying a filtering process to the trajectory in the coronal plane; The fourth information is Trajectories indicating the up-down, left-right, and front-back positions of the waist at each time point, which are restored from the trajectories obtained by applying filtering processing to the trajectories in the coronal plane and the transverse plane; The time for supporting with one leg on each side, From this, The fifth information is obtained by using the base frequency of the walking obtained as the lowest frequency peak by Fourier transforming the measurement data. Walking ability assessment device.

2. The computer a step of acquiring target data generated by observing the walking of a target person, the target data being configured to include at least any of first information indicating an area in the coronal plane of a locus of the center of gravity of the target person when walking, second information indicating an area in the transverse plane of the locus, third information indicating a degree of difference in the time during which the target person supports with each of the left and right legs when walking, fourth information indicating a degree of difference in the distance traveled while supporting with each of the left and right legs when walking, and fifth information indicating a degree of repetition of a walking action within a predetermined time; determining the walking ability of the subject based on the acquired subject data using a trained decision model generated by machine learning; outputting the result of determining the walking ability of the subject; Run the target data is configured to include all of the first information, the second information, the third information, the fourth information, and the fifth information, the target data is measurement data of an acceleration sensor attached to the waist of the target person, the first information is obtained by integrating the measurement data twice and plotting the left-right position and the up-down position at each time as a two-dimensional vector to obtain a coronal plane trajectory, and applying a filtering process to the coronal plane trajectory to calculate an area of the trajectory by integration; the second information is obtained by integrating the measurement data twice and plotting the left and right positions and the front and rear positions at each time as two-dimensional vectors to obtain a trajectory of the cross section, and applying a filtering process to the trajectory to calculate an area of the trajectory by integration; the third information is obtained from a time period during which the patient supports the body on one leg, the time period being calculated by measuring a time of a minimum point in the right vertical direction and a time of a minimum point in the left vertical direction of a trajectory obtained by applying a filtering process to the trajectory in the coronal plane; The fourth information is Trajectories indicating the up-down, left-right, and front-back positions of the waist at each time point, which are restored from the trajectories obtained by applying filtering processing to the trajectories in the coronal plane and the transverse plane; The time for supporting with one leg on each side, From this, The fifth information is obtained by using the base frequency of the walking obtained as the lowest frequency peak by Fourier transforming the measurement data. Methods for determining walking ability.

3. On the computer, a step of acquiring target data generated by observing the walking of a target person, the target data being configured to include at least any of first information indicating an area in the coronal plane of a locus of the center of gravity of the target person when walking, second information indicating an area in the transverse plane of the locus, third information indicating a degree of difference in the time during which the target person supports with each of the left and right legs when walking, fourth information indicating a degree of difference in the distance traveled while supporting with each of the left and right legs when walking, and fifth information indicating a degree of repetition of a walking action within a predetermined time; determining the walking ability of the subject based on the acquired subject data using a trained decision model generated by machine learning; outputting the result of determining the walking ability of the subject; Execute the target data is configured to include all of the first information, the second information, the third information, the fourth information, and the fifth information, the target data is measurement data of an acceleration sensor attached to the waist of the target person, the first information is obtained by integrating the measurement data twice and plotting the left-right position and the up-down position at each time as a two-dimensional vector to obtain a coronal plane trajectory, and applying a filtering process to the coronal plane trajectory to calculate an area of the trajectory by integration; the second information is obtained by integrating the measurement data twice and plotting the left and right positions and the front and rear positions at each time as two-dimensional vectors to obtain a trajectory of the cross section, and applying a filtering process to the trajectory to calculate an area of the trajectory by integration; the third information is obtained from a time period during which the patient supports the body on one leg, the time period being calculated by measuring a time of a minimum point in the right vertical direction and a time of a minimum point in the left vertical direction of a trajectory obtained by applying a filtering process to the trajectory in the coronal plane; The fourth information is Trajectories indicating the up-down, left-right, and front-back positions of the waist at each time point, which are restored from the trajectories obtained by applying filtering processing to the trajectories in the coronal plane and the transverse plane; The time for supporting with one leg on each side, From this, The fifth information is obtained by using the base frequency of the walking obtained as the lowest frequency peak by Fourier transforming the measurement data. Walking ability assessment program.

4. a data acquisition unit that acquires a plurality of learning data sets generated by observing the gait of each of a plurality of subjects, Each of the learning datasets is composed of a combination of training data and correct labels, the training data of each of the learning data sets is configured to include at least any of first information indicating an area in the coronal plane of a trajectory of the center of gravity of each of the subjects when walking, second information indicating an area in the transverse plane of the trajectory, third information indicating a degree of difference in time during which the subject supports with one leg each of the left and right legs when walking, fourth information indicating a degree of difference in distance traveled while supporting with one leg each of the left and right legs when walking, and fifth information indicating a degree of repetition of a walking action within a predetermined time period; The correct label of each of the learning data sets is configured to indicate a true value of the walking ability of each of the subjects. A learning data acquisition unit; a learning processing unit that performs machine learning of a judgment model using the acquired multiple learning data sets, the machine learning being configured by training the judgment model so that a result of judging the walking ability of each of the subjects by the judgment model based on the training data for each of the learning data sets conforms to a true value indicated by the correct answer label; Equipped with the training data is measurement data of an acceleration sensor attached to the waist of each of the subjects, the first information is obtained by integrating the measurement data twice and plotting the left-right position and the up-down position at each time as a two-dimensional vector to obtain a coronal plane trajectory, and applying a filtering process to the coronal plane trajectory to calculate an area of the trajectory by integration; the second information is obtained by integrating the measurement data twice and plotting the left and right positions and the front and rear positions at each time as two-dimensional vectors to obtain a trajectory of the cross section, and applying a filtering process to the trajectory to calculate an area of the trajectory by integration; the third information is obtained from a time period during which the patient supports the body on one leg, the time period being calculated by measuring a time of a minimum point in the right vertical direction and a time of a minimum point in the left vertical direction of a trajectory obtained by applying a filtering process to the trajectory in the coronal plane; The fourth information is Trajectories indicating the up-down, left-right, and front-back positions of the waist at each time point, which are restored from the trajectories obtained by applying filtering processing to the trajectories in the coronal plane and the transverse plane; The time for supporting with one leg on each side, From this, The fifth information is obtained by using the base frequency of the walking obtained as the lowest frequency peak by Fourier transforming the measurement data. Model generation device.

5. The computer A step of acquiring a plurality of training data sets generated by observing the gait of each of a plurality of subjects, Each of the learning datasets is composed of a combination of training data and correct labels, the training data of each of the learning data sets is configured to include at least any of first information indicating an area in the coronal plane of a trajectory of the center of gravity of each of the subjects when walking, second information indicating an area in the transverse plane of the trajectory, third information indicating a degree of difference in time during which the subject supports with one leg each of the left and right legs when walking, fourth information indicating a degree of difference in distance traveled while supporting with one leg each of the left and right legs when walking, and fifth information indicating a degree of repetition of a walking action within a predetermined time period; The correct label of each of the learning data sets is configured to indicate a true value of the walking ability of each of the subjects. Steps and A step of performing machine learning of a judgment model using the acquired multiple learning data sets, wherein the machine learning is configured by training the judgment model so that a result of judging the walking ability of each of the subjects by the judgment model based on the training data for each of the learning data sets matches a true value indicated by the correct answer label; Run the training data is measurement data of an acceleration sensor attached to the waist of each of the subjects, the first information is obtained by integrating the measurement data twice and plotting the left-right position and the up-down position at each time as a two-dimensional vector to obtain a coronal plane trajectory, and applying a filtering process to the coronal plane trajectory to calculate an area of the trajectory by integration; the second information is obtained by integrating the measurement data twice and plotting the left and right positions and the front and rear positions at each time as two-dimensional vectors to obtain a trajectory of the cross section, and applying a filtering process to the trajectory to calculate an area of the trajectory by integration; the third information is obtained from a time period during which the patient supports the body on one leg, the time period being calculated by measuring a time of a minimum point in the right vertical direction and a time of a minimum point in the left vertical direction of a trajectory obtained by applying a filtering process to the trajectory in the coronal plane; The fourth information is Trajectories indicating the up-down, left-right, and front-back positions of the waist at each time point, which are restored from the trajectories obtained by applying filtering processing to the trajectories in the coronal plane and the transverse plane; The time for supporting with one leg on each side, From this, The fifth information is obtained by using the base frequency of the walking obtained as the lowest frequency peak by Fourier transforming the measurement data. Model generation method.

6. On the computer, A step of acquiring a plurality of training data sets generated by observing the gait of each of a plurality of subjects, Each of the learning datasets is composed of a combination of training data and correct labels, the training data of each of the learning data sets is configured to include at least any of first information indicating an area in the coronal plane of a trajectory of the center of gravity of each of the subjects when walking, second information indicating an area in the transverse plane of the trajectory, third information indicating a degree of difference in time during which the subject supports with one leg each of the left and right legs when walking, fourth information indicating a degree of difference in distance traveled while supporting with one leg each of the left and right legs when walking, and fifth information indicating a degree of repetition of a walking action within a predetermined time period; The correct label of each of the learning data sets is configured to indicate a true value of the walking ability of each of the subjects. Steps and A step of performing machine learning of a judgment model using the acquired multiple learning data sets, wherein the machine learning is configured by training the judgment model so that a result of judging the walking ability of each of the subjects by the judgment model based on the training data for each of the learning data sets matches a true value indicated by the correct answer label; Execute the training data is measurement data of an acceleration sensor attached to the waist of each of the subjects, the first information is obtained by integrating the measurement data twice and plotting the left-right position and the up-down position at each time as a two-dimensional vector to obtain a coronal plane trajectory, and applying a filtering process to the coronal plane trajectory to calculate an area of the trajectory by integration; the second information is obtained by integrating the measurement data twice and plotting the left and right positions and the front and rear positions at each time as two-dimensional vectors to obtain a trajectory of the cross section, and applying a filtering process to the trajectory to calculate an area of the trajectory by integration; the third information is obtained from a time period during which the patient supports the body on one leg, the time period being calculated by measuring a time of a minimum point in the right vertical direction and a time of a minimum point in the left vertical direction of a trajectory obtained by applying a filtering process to the trajectory in the coronal plane; The fourth information is Trajectories indicating the up-down, left-right, and front-back positions of the waist at each time point, which are restored from the trajectories obtained by applying filtering processing to the trajectories in the coronal plane and the transverse plane; The time for supporting with one leg on each side, From this, The fifth information is obtained by using the base frequency of the walking obtained as the lowest frequency peak by Fourier transforming the measurement data. Model generator.

Images